CN110475697A - System and method for calibrating vehicle tyre - Google Patents
System and method for calibrating vehicle tyre Download PDFInfo
- Publication number
- CN110475697A CN110475697A CN201780088545.4A CN201780088545A CN110475697A CN 110475697 A CN110475697 A CN 110475697A CN 201780088545 A CN201780088545 A CN 201780088545A CN 110475697 A CN110475697 A CN 110475697A
- Authority
- CN
- China
- Prior art keywords
- state
- vehicle
- rigidity
- track
- movement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims description 76
- 238000009826 distribution Methods 0.000 claims abstract description 83
- 238000005259 measurement Methods 0.000 claims abstract description 40
- 230000008859 change Effects 0.000 claims abstract description 13
- 239000002245 particle Substances 0.000 claims description 31
- 230000003993 interaction Effects 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 4
- 238000009877 rendering Methods 0.000 claims description 2
- 230000001052 transient effect Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 12
- 238000005070 sampling Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000000418 atomic force spectrum Methods 0.000 description 6
- 239000004744 fabric Substances 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000014509 gene expression Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 2
- 241001061260 Emmelichthys struhsakeri Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000003321 amplification Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 240000001439 Opuntia Species 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T8/00—Arrangements for adjusting wheel-braking force to meet varying vehicular or ground-surface conditions, e.g. limiting or varying distribution of braking force
- B60T8/17—Using electrical or electronic regulation means to control braking
- B60T8/172—Determining control parameters used in the regulation, e.g. by calculations involving measured or detected parameters
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/12—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to parameters of the vehicle itself, e.g. tyre models
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T2210/00—Detection or estimation of road or environment conditions; Detection or estimation of road shapes
- B60T2210/10—Detection or estimation of road conditions
- B60T2210/12—Friction
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T2270/00—Further aspects of brake control systems not otherwise provided for
- B60T2270/86—Optimizing braking by using ESP vehicle or tyre model
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2400/00—Indexing codes relating to detected, measured or calculated conditions or factors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/26—Wheel slip
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/28—Wheel speed
Landscapes
- Engineering & Computer Science (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Tires In General (AREA)
Abstract
The tire of vehicle is calibrated by the measurement model associated with vehicle-state of the motion measure by vehicle of motion model associated with vehicle-state and vehicle is inputted to the control of vehicle based on vehicle.The motion model of vehicle includes the combination of the certainty component of movement and the probability component of movement, wherein the certainty component of movement is independently of rigidity state and defines the movement that vehicle changes over time and change.The probability component of movement includes the rigidity state with uncertain and definition to the interference of vehicle movement.The probability distribution of rigidity state is iteratively updated until meeting termination condition, calibrate tire based on exercise data, exercise data includes the sequence of the measured value for the movement for moving vehicle along track according to the sequence of the mobile control input to vehicle in track and vehicle.Iteration determines the first state track of vehicle using one or more samples of the sequence of control input and the probability distribution of rigidity state according to motion model, the second state trajectory of vehicle is determined according to measurement model using the sequence of measured value, and updates the probability distribution of rigidity state to reduce the error between the first state track of vehicle and the second state trajectory of vehicle.
Description
Technical field
The present invention relates to the tire of vehicle-road interactions, and more particularly, to the ginseng of calibration vehicle tyre
Number.
Background technique
Tire-road interaction is to generate or change the principal element of wheeled vehicle movement, and know and tire-road
The road relevant variable that interacts is necessary for many active safety systems in modern vehicle.In many modern vehicles
Parameter relevant to road friction is used in.For example, anti-lock braking system (ABS), electronic stabilizing control system (ECS) and
Senior Officer's auxiliary system (ADAS) use parameter relevant to tire-road interaction all expansiblely, to mention
For advanced security mechanism.
Important parameter when determining tire-road interaction is coefficient of friction.The coefficient of friction known is used as
For the supervision component of driver, but it also can be used in such as ABS, ESC and ADAS.Coefficient of friction can be used directly
In vehicle control system as such as ADAS；It is used as the supervision component for driver, for example, warning driver is on road
There is sudden change in face；It can be used for the road surface for just travelling automobile on it classification.
Another parameter of tire stiffness state is power-curve of sliding initial slope.Tire stiffness in forward direction and
It is usually different in transverse direction, therefore each wheel is likely to the individually power-curve of sliding of depending therefrom two.Wheel
Tire rigidity can be used directly in vehicle control system as such as ADAS；It is used as the supervision component for driver,
For example, warning driver has sudden change in road surface；It can be used for the road surface for just travelling automobile on it classification；And/or
It can be used to determine coefficient of friction.
Therefore, it is desirable to know friction and/or can help to determine the other parameters of friction when driving.For example, with tire-
The relevant parameter of power that road contact generates is sliding.In longitudinal situation, that is, on the forward direction of wheel, foundation
Returned by the rotation speed of wheel or longitudinal rate (no matter wherein which is bigger, that is, no matter wheel is to accelerate or braking)
Longitudinal rate of one wheel changed and the difference of rotation speed are slided to define.In lateral situation, that is, in the transverse direction side of wheel
Upwards, sliding is defined according to the ratio between the lateral velocity component of vehicle and longitudinal velocity component.
It is however typically difficult to directly measure or sense during driving such as peak value coefficient of friction, tire stiffness and sliding
Such parameter relevant to tire-road interaction；Therefore, usually by between one or more sensor uses of combination
It connects friction and determines method to determine the parameter.
Many methods are intended to estimate using various optimisation techniques the parameter of tire.For example, US8, described in 065,067
Method carrys out Nonlinear Function Approximation using the data bins being collected into, and makes the mistake of friction and tire stiffness using nonlinear optimization
Difference minimizes.It is well known, however, that nonlinear optimization often lacks convergence in local optimum or often restrains.
In addition, all methods for being intended to determine tire parameter during runing time both depend on good initial guess.School
The method of quasi- tire parameter is typically based on high-precision sensor setting or testboard.However, high-precision sensor is at high cost, and
Testboard is approximate real world, so that the calibration of specific tires will be only applicable to the specific testboard.
Available sensor is when therefore, it is necessary to use production vehicle the system and method for calibrating vehicle tyre parameter.
Summary of the invention
The purpose of some embodiments of the present invention is to provide the system for determining vehicle tyre parameter, the System describe
Tire-road contact power relationship.Another purpose of some embodiments, which is to provide, to be suitable for use in determining in standard passenger vehicle
This method of rigidity state when available low cost sensor.As defined herein, the parameter of tire includes defining vehicle
At least one parameter of the interaction of road for just travelling on it of at least one tire and vehicle.For example, parameter can
Friction between the lateral stiffness of longitudinal rigidity, tire including tire, tire and road determines that tire and road contact power close
One or combinations thereof in the form factor of the shape of system.
Another purposes of some embodiments be probability determine the parameter of tire and/or estimate that determined parameter is set
Reliability.For example, an embodiment is initialized using the confidence level of rigidity state for the change according to road surface and tire pressure
Change the method for adjusting the calibration parameter of tire.As used in this article, vehicle can be such as car, bus or rover station
Such any kind of wheeled vehicle.
Some embodiments are the understanding that the movement based on vehicle depends on tire parameter.For this purpose, attempting by using receipts
The sensing data set collected estimates the state trajectory of vehicle iteratively to estimate the parameter of tire, and estimating using vehicle
The state trajectory and motion model of meter update the parameter of tire.
The problem of this method, is that the time-evolution of rigidity state is unknown, therefore the movement including rigidity state
Model is unknown and cannot be verified.However, some embodiments are based on the insight that unknown rigidity state can be regarded
To act on random perturbation of other deterministic models of vehicle movement to generate the motion model with unknown rigidity.
On the other hand, the measurement model of vehicle is also denoted as including rigidity state, such as, it may include longitudinal rate,
The speed of rotation of lateral rate and vehicle.In this way, rigidity can be indicated at least indirectly with motion model and measurement model
State.Specifically, movement and measurement model by being associated by the state trajectory of vehicle traction, if rigidity state it is known that if
The state trajectory should be identical.The variation of difference hinders the determination of rigidity value, but allows to determine the general of rigidity state
Rate distribution.
Some embodiments are other understanding that the Probability State based on rigidity is not suitable for auto model.However, some realities
The mode of applying, which is based on the insight that, can adopt the feas ible space of the parameter of the rigidity state defined by its probability distribution
Sample, and sampling parameter is used in Combined estimator rigidity state.However, Combined estimator should be not used for updating adopting for rigidity
Sample state, and be used to update the probability distribution of rigidity state.For example, can be distributed with update probability, to reduce two states
The value of rigidity state used in the successive iterations that error and influence between track extract in the distribution from update.
Therefore, an embodiment discloses a kind of for estimating the rigid of vehicle by using the vehicle-state track of estimation
Degree state is come the method for calibrating vehicle tyre, wherein vehicle-state includes the rate and course angular speed (heading of vehicle
Rate), and wherein, the parameter of tire includes the road for defining at least one tire of vehicle and just travelling on it with vehicle
At least one parameter of interaction.This method includes will input and vehicle to the control of vehicle from memory search vehicle
The measurement model of state associated motion model and vehicle, wherein the motion model of vehicle be movement certainty component and
The combination of the probability component of movement, and define the movement that vehicle changes over time and changes, wherein probability motion model includes
With uncertain and define the parameter of the interference to vehicle movement, wherein measurement model is by measured value and vehicle-state phase
Association；The probability distribution of rigidity state is iteratively updated until meeting termination condition；Update rigidity state probability distribution so that
Error between the first state track of vehicle and the second state trajectory of vehicle reduces；Export rigidity state probability distribution and
Represent at least one of rigidity state estimation of the probability distribution of rigidity state or combinations thereof.It is deposited using being operably connected to
The step of at least one processor of reservoir executes this method.
Another embodiment discloses a kind of rigidity shape for the vehicle-state track estimation vehicle by using estimation
State is come the system of calibrating vehicle tyre, wherein vehicle-state includes the rate and course angular speed of vehicle, and wherein, tire
Parameter include define vehicle at least one tire and the interaction of road that is just travelling on it of vehicle at least one
Parameter.The system includes: memory, and storage vehicle will input movement associated with the state of vehicle to the control of vehicle
The measurement model of model and vehicle, wherein the motion model of vehicle is the certainty component of movement and the probability component of movement
Combination, and defines the movement that vehicle changes over time and changes, wherein probability motion model include with uncertain and
Define the parameter of the interference to vehicle movement, wherein the measurement model of vehicle is associated with measured value by the movement of vehicle；Processing
Device is used to update the probability distribution of rigidity state until meeting termination condition；Output device, for rendering rigidity state
Probability distribution and represent one of rigidity state estimation of probability distribution of rigidity state or combinations thereof.
Another embodiment discloses a kind of non-transient computer readable storage medium of implementation procedure on it, the journey
Sequence can be executed by processor to execute a kind of method.This method includes from memory search vehicle the control to vehicle is defeated
Enter related with the state of vehicle to the measured value of the movement by vehicle of the associated motion model of the state of vehicle and vehicle
The measurement model of connection, wherein the motion model of vehicle includes the combination of the certainty component of movement and the probability component of movement,
In, the certainty component of movement is independently of rigidity state and defines the movement that vehicle changes over time and change, wherein movement
Probability component include the rigidity state with interference that is uncertain and defining movement to vehicle；Receive instruction vehicle root
The exercise data moved on road according to track, wherein exercise data includes the control to vehicle for keeping vehicle mobile according to track
The sequence of the measured value of the movement of vehicle making the sequence of input and moving along track, and wherein, the sequence pair of measured value
It should be in the sequence of control input；The probability distribution of rigidity state is iteratively updated until meeting termination condition, wherein iteration uses
The one or more samples for controlling the sequence of input and the probability distribution of rigidity state determine the of vehicle according to motion model
One state trajectory, the second state trajectory of vehicle is determined using the sequence of measured value according to measurement model, and updates rigidity
The probability distribution of state reduces the error between the first state track of vehicle and the second state trajectory of vehicle；And
At least one of the sample of probability distribution of the probability distribution and the rigidity state when meeting termination condition of rigidity state is presented
Or combinations thereof.
Detailed description of the invention
[Figure 1A]
How Figure 1A is the size of the power on vehicle tyre driving on the road with for different types of link table
The sliding in face and the illustration changed.
[Figure 1B]
How Figure 1B is the size of the power on vehicle tyre driving on the road with for different types of tyre mould
The longitudinal sliding motion of type and the illustration for sliding laterally the two variation.
[Fig. 1 C]
Fig. 1 C is the amplification form of Figure 1A.
[Fig. 1 D]
Fig. 1 D is how by approaching the curve with a variety of rigidity states to approach the illustration of tire force curve.
[Fig. 1 E]
Fig. 1 E is times of the rigidity state for iteratively estimating vehicle tyre according to embodiment of the present invention
The flow chart of several methods.
[Fig. 1 F]
Fig. 1 F is the probability distribution for showing the feas ible space of definition rigidity state according to embodiment of the present invention
The curve graph of function.
[Fig. 1 G]
Fig. 1 G be for illustrate embodiment according to the present invention use principle pass through it is general with weighted state trajectory creation
Rate distribution function determines the curve graph of the method for first state track.
[Fig. 2A]
Fig. 2A is the flow chart for the method that the principle that embodiment according to the present invention uses generates first state track.
[Fig. 2 B]
Fig. 2 B is the stream of the method for the probability distribution that the principle that embodiment according to the present invention uses updates rigidity state
Cheng Tu.
[Fig. 3 A]
Fig. 3 A is the schematic diagram of the common tire calibrator of tire.
[Fig. 3 B]
Fig. 3 B is according to one embodiment for determining or estimating the block diagram of the method for the internal signal from vehicle.
[Fig. 3 C]
Fig. 3 C is according to one embodiment by the curve graph of rigidity state estimated by curve matching.
[Fig. 4 A]
Fig. 4 A is the schematic diagram of simplified auto model.
[Fig. 4 B]
Fig. 4 B is the schematic diagram of full custom auto model (full-order vehicle model).
[Fig. 5 A]
Fig. 5 A is the signal for the probability for how specifying some principles that embodiment uses according to various embodiments of the present invention to determine
Figure.
[Fig. 5 B]
Fig. 5 B is the signal for how specifying probability that some principles used according to various embodiments of the present invention determine
Figure.
[Fig. 6 A]
Fig. 6 A is to generate first state when for five sampling rigidity values of each grey iterative generation according to some embodiments
The schematic diagram of the result of the iteration three times of track.
[Fig. 6 B]
The possibility of five kinds of states specifies the schematic diagram of probability when Fig. 6 B is first time iteration in Fig. 6 A.
Specific embodiment
How Figure 1A shows the size of the power on the vehicle tyre travelled on road surface with for such as dry drip
Blueness 110, wet pitch 120 and snow 130 surfaces as different types of road surface sliding and change.Tire force relationship is height
It is nonlinear, and additionally depend on the abrasion on such as tire pressure, vehicle mass, tyre temperature and tire it is such other
Amount.As used in this article, vehicle can be any kind of wheeled vehicle as such as car, bus or rover station
.
Figure 1A shows example scenario when the every other amount in addition to sliding is kept fixed.This is to illustrate tire force
Method of relationship itself.Figure 1A can illustrate longitudinal force, in the case where longitudinal force, according to the rotation speed or longitudinal direction for passing through wheel
Longitudinal rate of rate (no matter wherein which is bigger) normalized wheel and the difference of rotation speed are slided to define.Figure 1A can
Cross force is illustrated, in the case where cross force, is come according to the ratio between the lateral velocity component and longitudinal velocity component of vehicle
Definition sliding.
Figure 1A shows the situation when only one slippage non-zero simultaneously.In general, for example, the driver when vehicle exists
When also rotating while braking, two slippages are all non-zeros.According to used specific tires model, power-sliding relation
Seem to be different.
How the power that Figure 1B shows the vehicle tyre when the variation of the value of slippage changes.This scenario show two kinds of differences
The result of tire model: weighted function model 110b and friction model of ellipse 120b.In remaining description of the invention, for letter
For the sake of list, one of slippage is assumed zero, it is to be appreciated that and as with experience in the art it will be appreciate that,
The present invention described herein also contemplated the case where combination sliding.
Fig. 1 C shows the amplification form of Figure 1A, and wherein power carries out normalizing by the normal force being shelved on wheel
Change, wherein the case where considering dry pitch 110 in more detail.The value that power reaches its maximum value is referred to as peak value frictional force 112.Peak
Value frictional force 112 is known in several automotive control systems.For example, the knowledge of peak value frictional force can apply understanding
The braking torque of specific wheel in electronic stability controlling system (ESC) is how much to be important.Peak value friction valve and phase
The slip value 113 answered can be used for anti-lock braking system (ABS) to realize optimum braking force.The initial slope 111 of force curve 110
The commonly known as rigidity of tire.During normal driving (sliding in this case small), it can be approached with tire stiffness 111
Force curve.
Relatively, Fig. 1 D shows the approximate model of tire force curve used in several embodiments in the present invention.
In this particular example, with the interval corresponding four sections of affine curve 111d, 112d of 101d, 102d, 103d and 104d, 113d,
114d approaches force curve 110.That is, the linearity curve of approaching to reality force curve is different according to the value of slip value.Such as
What subsequent implementation mode of the invention was explained, estimate a plurality of linearity curve compared estimate by the single non-thread of different constant parameters
Property function is easier.For example, the normal method modeled to tire force relationship is the Magic by being given by
Formula or Pacejka model
F0(m)=D sin (C arctan (Bm-arctan (Bm)))) (1)
Wherein, B is stiffness factor, and C is form factor, and D corresponds to the peak factor of peak value coefficient of friction, and E is curvature
The factor, F0It is longitudinal force or cross force, and m is longitudinal sliding motion or slides laterally.Formula (1) is nonlinearity and depends on
In 5 parameters.However, because cannot direct measuring force, be difficult to determine formula (1) in parameter.Alternatively, if using dividing
Section is affine to approach, then can use
Fx≈Cλλ
Fy≈CαThe a plurality of linearity curve of alpha form replaces non-linear relation (1),
Wherein, CλIt is the slope of linearity curve, the tire stiffness on wheel longitudinal direction, CαIt is in wheel transverse direction
Tire stiffness, λ is longitudinal wheel skid, and α is lateral wheel skid.
Fig. 1 E shows the stream of method of the probability distribution for iteratively updating rigidity state until meeting termination condition
Cheng Tu.The embodiment is based on the insight that unknown rigidity state can be considered as acting on other certainty of vehicle movement
The random perturbation of model.The property of random perturbation causes vehicle that there is different possibility to move, therefore has different possibility shapes
State.
For this purpose, the embodiment is retrieved from memory as the vehicle relevant control for vehicle of vehicle-state inputs
Movement and vehicle relevant to the measurement of the vehicle movement with vehicle-state measurement model.The motion model of vehicle includes
The combination of the probability 112e component of the certainty component 111e and movement of movement, wherein the probability component of movement includes having not
The interference of deterministic rigidity state and definition to vehicle movement.Then, this method indicates road vehicle root in storage
Retrieval 120e is carried out in the memory of the exercise data 119e from vehicle testing driving of the movement carried out according to track, wherein
Exercise data includes the sequence for making to be moved along the mobile control input 130e for vehicle in indicated track.Exercise data is also
The sequence of the motion measure of the vehicle moved including 135e along track, and wherein, the sequence of measured value corresponds to control
The sequence of input.Then, this method using control input 130e and probability component 111e and probability component 112e combination and
Definition is to one of rigidity probability sample states of interference of vehicle movement or a variety of, indicate vehicle movement with formation first
State trajectory 150e.In addition, this method determines the second state trajectory using measured value 135e and measurement model 145e, and more
The probability distribution of new 160e rigidity state, to reduce the mistake between the first state track of vehicle and the second state trajectory of vehicle
Difference.If meeting termination condition 170e, this method exports the probability distribution and rigidity shape probability of state point of 180e rigidity state
One or combinations thereof in the sample of cloth.
In some embodiments, measurement model includes independently of the certainty component of rigidity state and including rigidity state
Measurement model probability component combination.In one embodiment, the combination be used for using measured value sequence and just
The sample of degree state determines the second state trajectory.
Fig. 1 F, which is shown, illustrates the curve graph of the probability-distribution function 140f for the feas ible space 100f for defining rigidity state.It can
To predefine the shape of function 140f, that is to say, that the expression of the state of rigidity 140e.For example, if point of rigidity state
Cloth is Gaussian Profile, then the shape for being distributed 140f is " Gaussian hat " shape.If shape be it is fixed, in the feelings of Gaussian Profile
Under condition, average value 110f and variance 130f define distribution 140f and therefrom extract the feas ible space of the sample parameter of rigidity state
100f.However, the expression of many other rigidity states can be used.For example, if using Normal-inverse Wishart points
Cloth, then sufficient statistical data includes four parameters, and wherein mean value is one of them.In general, any expression can be used, still
Some expressions of rigidity state are more suitable for certain motion models.One embodiment is right from each of control input and measured value
One of rigidity state or a variety of is extracted in the probability distribution for the rigidity state that should be worth.For example, in one embodiment, just
The probability distribution of degree state is modeled as the Gaussian Profile defined by its mean value and variance, and with control input and measured value
The corresponding each step of time step in, from one or more of sample drawn in distribution and be used as in motion model
Component.Therefore, by extracting several samples, each control input will obtain slightly different track, which will be with measured value
Sequence in the second state trajectory it is more or less consistent.By keeping redundancy sampling value when generating first state track,
The number of iterations of this method can be reduced.
As used in this article, carrying out sampling to the value of rigidity state is divided with being distributed the probability decimation value that 140f is defined
Cloth 140f is defined by the underlying parameter of rigidity state again.For example, according to distribution 140f, the probability that sample 120f is extracted or samples
Higher than the probability of sample 150f.This probability distribution for indicating to allow iteratively to update 160f rigidity state, is divided with generating to update
Cloth 145f, update distribution 145f is defined in the successive iterations before meeting termination condition 170e to be used to sample rigidity by what is used
The update feas ible space of state.
G referring to Fig.1, in one embodiment, this method passes through to be inputted with by a variety of rigidity sample states and control
It is generating, by constructing probability point to several state trajectory 120g of the corresponding consensus weighting of the second state trajectory state trajectory
Cloth function 110g and state trajectory 130g is extracted from distribution 110g generate first state track.Begin in this way
Ensure that the first state track for updating sufficient statistic evidence can indicate the distribution of state trajectory eventually.
The fact that rigidity state is relative to each other with control input by the power of tire is utilized in several embodiments.At one
In embodiment, control input includes one or combinations thereof in the steering angle of designated vehicle wheel and the rotation speed of wheel
The order of value.The steering angle influence of wheel can be in power 110b, 120b that either wheel generates upwards.For example, if wheel
Steering angle be zero, then it is mobile to be along line for vehicle, and power will be directed toward a direction.However, if enforcing non-
Zero steering angle, then power will be divided into longitudinal component and cross stream component, so that how will affect can determine rigidity state.One implementation
Mode uses the measured value of one or combinations thereof value in the speed of rotation and vehicle acceleration including vehicle, it means that right
The measurement model answered will have depending on the component of rigidity state and independently of the one-component of rigidity state.Therefore, a reality
Applying mode utilizes the information about rigidity state to be incorporated directly into the fact in the sequence of measured value.State trajectory can be with more
Kind mode is constituted.In one embodiment, state trajectory includes status switch, and every kind of state all includes the rate and boat of vehicle
To angular speed, so that motion model controls the first value of the value vehicle-state of input with continuous time step, and measurement model is with phase
Same time step is associated with the second value of vehicle-state by measured value.
Generating first state track 150e can carry out in a number of ways.Fig. 2A shows some implementations according to the present invention
The block diagram of the method for the generation first state track 150e of mode.The processor of vehicle can be used to implement this method.This method is adopted
Sample 250 is by the feas ible space for the possibility rigidity state being used in the motion model 245 of vehicle, to generate the rigidity state sampled
Set.Due to the uncertainty of rigidity state, the motion model 245 of vehicle includes the uncertainty of vehicle movement.This method makes
Motion model 245 with vehicle and the input 247 to vehicle determine 260 1 groups of particles 265, this group of particle 265 represent due to
Different modes influence may movement caused by the different sample states of the rigidity 255 of vehicle movement.For control and measured value
Each step of sequence carries out the determination of particle, to form the track for representing the particle of first state track.
Next, this method with measurement relationship 266 in the consistent each step of measured value in use identified grain
Son 265, and with from either on vehicle or being connected remotely to the second of the vehicle that the memory of vehicle processor obtains
State trajectory 267 is compared, and how to be possible to indicate the weight 270 of time of day with each particle of update instruction.Measure mould
Identified each particle is mapped to identified measuring state by type 266, and this method is by calculating the second state trajectory
Difference between 267 and identified each measuring state generates updated weight 275, with each grain determined by reflecting
How son 265 reflects measured value.If the end of the sequence of measured value has not yet been reached, that is to say, that if in exercise data simultaneously
Not all measured value all has been subjected to processing, then this method carries out sampling 250 to feas ible space repeatedly.If arrived measured value
The end of sequence, then this method determines the distribution of 280 state trajectories.For example, limited weight sets 275 can be used directly in method 280
To reflect that distribution or this method of state trajectory can generate state by using such as cuclear density smoother smoothes weight
Track it is continuously distributed.Then, first state track 290 is generated using obtained distribution 285.For example, an embodiment
First state track 295 is generated by generating sample with distribution 285.First state track is determined as by another embodiment
According to the combination of the original state track of the Weight of each initial first state track.In yet another embodiment, one
Bar state track remains in a particle.For example, by using vehicle movement modeling system and can ensure this
State trajectory, which is kept always as one of particle, pre-defines the state trajectory.Alternatively, being utilized in each determining particle
One of state trajectory and particle are connect by the previous iteration of this method, wherein by with directly proportional to updated weight 275
The value of one of the particle that probability indexes particle is sampled and is attached.In some embodiments, in addition to state trajectory
Except, the entity of output includes the sample of the feas ible space generated for generating state trajectory.
Although be mathematically it is equivalent, to select which of them may be to the practical implementation of method
Influence it is big, for example, this is because limited numerical accuracy in digital processing unit executes calculates either because of input and state
Digital zooming difference between track.
Fig. 2 B shows the block diagram of an iteration of the method for the probability distribution for updating rigidity state.This method determines
Indicate the error locus 210b of the difference between first state track 209b and the second state trajectory 206b.This method uses vehicle
The measurement model of 207b, the motion model of vehicle 208b and n error locus 211b is generated to the input of vehicle.Then, should
Method updates 220b probability distribution by using error locus 211b.In some embodiments, by the previous of this method
The abundant statistics and error locus estimated in iteration are averaging to be updated 220b.
In some embodiments, by the error indicated between initial first state track and the second state trajectory
The first state track 295 that each initial first state track is weighted and generates is used by reduction first state track
Carry out direct update probability distribution with the error obtained between the movement of first state track when being used in motion model.This error
The error due to caused by the error of rigidity state is reflected, and can be used to update the probability distribution of rigidity state.This is
It is possible, because the second state trajectory has been used for determining first state track and is influencing each first state track
Weight.
Fig. 3 A shows the overall structure by the system 399 for estimating the state calibration vehicle tyre of vehicle tyre rigidity,
The rigidity state includes defining at least one parameter of the interaction of at least one tire and road of vehicle.Tire calibrator
399 include (such as, iteratively updating the probability distribution of rigidity state until meeting for executing the module of tire calibrator 399
Termination condition) at least one processor 307, wherein iteration using control list entries and rigidity state probability distribution
One or more samples determine the first state track of vehicle according to motion model, using the sequence of measured value according to measurement mould
Type determines the second state trajectory of vehicle, and updates the probability distribution of rigidity state to reduce the first state track of vehicle
Error between the second state trajectory of vehicle.Memory 380 is arrived in the connection of processor 370 371, and memory 380 stores vehicle
The measurement by vehicle movement that the control to vehicle is inputted to associated with the state of vehicle motion model 381 and vehicle
It is worth measurement model 382 associated with vehicle-state, wherein the motion model of vehicle includes certainty component and the movement of movement
Probability component combination, wherein the certainty component of movement independently of rigidity state and define vehicle change over time and
The movement of variation, wherein the probability component of movement includes the rigidity with uncertain and definition to the interference of vehicle movement
State.
Calibrator can also store the exercise data of movement of the 383 instruction vehicles on road according to track, wherein movement number
According to including making vehicle according to the sequence of the mobile control input to vehicle in track and the movement of the vehicle moved along track
The sequence of measured value, and wherein, the sequence of measured value corresponds to the sequence of control input.Alternatively, calibrator may include
Receiver 390, to receive the exercise data 369 of movement of the instruction vehicle according to track on road.The system further includes output
350 device of device, the probability distribution and the probability distribution of the rigidity state when meeting termination condition of rigidity state is presented
At least one of sample or combinations thereof.In order to realize certain embodiments of the invention, memory 380 also stores 383 estimations
The calculated state trajectory of each of the vehicle of the internal information of device, the including but not limited to value of rigidity state, different iteration
It is worth, causes the movement of every kind of state of vehicle and leads to the feas ible space of the sampling of state trajectory.
In one embodiment, before execution, the exercise data of instruction vehicle movement is pre-processed.Fig. 3 B shows root
According to embodiment for determining or the method for the internal signal from vehicle that estimation will use by tire calibrator
Block diagram.The step of leading to 381b to 361b, can realize that the control unit or circuit arrangement can in control unit or circuit arrangement
It is used in system as such as ABS, ESP, ADAS or in autonomous vehicle.For example, input signal filter 310b can pass through
The speed of rotation of wheel or tire 309b are handled to determine input signal to generate signal 311b, thus can be directed to each of vehicle
Independent wheel or tire determine the speed of rotation.Filter 310b can also determine input signal by processing brake pressure 208b
312b, and input signal 313b is determined by handling the speed of rotation from engine 307b and torque.Block 330b is determined
Longitudinal acceleration 331b, and brake force estimator 340b is directed to the braking of each wheel using the brake pressure 313b estimation applied
Power 341b.According to the value of motor torque and the speed of rotation 314b of engine, the module in control unit estimates longitudinal direction
On driving force, and vertical force 351b is for example estimated in 350b using the estimated value of longitudinal acceleration 331b.
Use vertical force estimated value 351b and longitudinal force estimated value 341b and 371b, it may be determined that normalization longitudinal force 361b.
Radius of wheel estimator 320b is estimated using the processed speed of rotation 311b's of tire or wheel and normalization driving force 361b
Evaluation carrys out correction wheel radius, and radius of wheel is exported together with speed of rotation 321b.For example, radius of wheel estimator
320b estimated wheel slides 321b.Therefore, signal conditioner 320 can provide longitudinal velocity for rigidity state estimator 340
The estimated value of 321b, wheel skid estimated value 381b or normalization longitudinal force 361b or combinations thereof.Therefore, in some embodiments
In, tire calibrator 350 uses one in longitudinal rate 321b, wheel skid estimated value 381b and normalization longitudinal force 361b
Or combinations thereof estimated value.
Wheel aligner had both been used as or the initializer for the estimator that rubs in real time or own represent calibration
The method of tire parameter.For example, looking back the formula (1) as the possibility model of relationship between sliding and friction, there are five different
Parameter will be estimated.For this purpose, referring to Fig. 3 V, parameter of an embodiment of the invention in optimization process in estimator (1), In
In optimization process, the output of formula (1) and rigidity state estimation 310 difference between c obtain the formula (1) as in 311c.
In some embodiments, vehicle-state is according to the motion model of vehicle-state dynamic evolution in time.For example,
It can be according to certain nonlinear functionTo describe the motion model of vehicle, wherein v is random
Variable, i.e. process noise.In some embodiments, process noise is the basis with unknown mean μ and variance ∑Gauss, and wherein, process noise vkWith measurement noise ekIt may depend on each other.In addition, f (xk) it is to retouch
State the nonlinear deterministic function of vehicle-state differentiation.Similarly, g (xk) it is by stochastic variable or interference map to vehicle-state
Nonlinear deterministic function, xkIt is state and k is time index.
The dynamic model of vehicle movement depends on the state of rigidity, the variance of mean value, rigidity including rigidity and each wheel
The coefficient of friction in each direction of tire.In some embodiments, the disturbance v of vehicle movement is influencedkIt is due to tire stiffness
Number description in uncertainty.In other embodiments, the state of vehicle includes the rate vector and course angle of vehicle
Rate.
Different types of motion model can be used.For calculation purposes, naive model is preferably as it facilitates fastly
Speed, which is realized, to be calculated.But if model accuracy is important, preferably high-fidelity model.In addition, the mould used according to the present invention
Type, the parameter of adjustable different number.For example, Fig. 4 A shows the schematic diagram of the simplification front-wheel drive single track model of vehicle,
In two wheels on each axis be brought together.The model depends on four kinds of rigidity states when accelerating, a longitudinal direction two
A transverse direction.In fig. 5, δ is the steering angle of front-wheel, and α is the sliding in lateral situation, and β is the vehicle body sliding of vehicle, is determined
The ratio of the adopted forward direction rate for vehicle and lateral rate, and FX, yIt is longitudinal (forward direction) and cross force respectively.
Fig. 4 B shows the schematic diagram of double track model, all four wheels are all modeled.Using this model, 8 kinds just
The movement of degree state influence auto model.In the following embodiments, the single track model in Fig. 5 A is used as model, but to recognize
To the model that can easily use in Fig. 5 B.Fig. 4 B shows the schematic diagram of the full chassis model with front steering.This
In the case of, the number of parameter to be estimated increases, it is to be appreciated that application is identical method.
In order to illustrate why state trajectory give about estimation rigidity state when accuracy information, Fig. 5 A illustrate
Vehicle has the scene of original state 510.For the one group of sampling rigidity state sampled from the probability distribution of rigidity state
And the given input to system, vehicle is obeyed movement 511a and is terminated in 530a, so as to cause uncertain 531a.By
Intrinsic uncertainty in second state trajectory caused by biasing and remaining sensor error in noise, sensor is led
It causes until specific region 520 can only known vehicle-state.However, the end-state of vehicle 530a is located at region well
In 520, therefore the specific combination of this rigidity state and vehicle original state is endowed high good combination probability.Therefore, rigidity
The probability distribution of state is likely to be good distribution.
Fig. 5 B show with identical original state 510, may sensor bias term having the same but have another
The vehicle of the specific rigidity state of group.For the identical input to system, vehicle 510 obeys movement 511b at this time, to cause
Vehicle terminates in state 530b, leads to uncertain 531b.However, this of vehicle, which terminates state 530b, is not located at sensor really
Determine in region.Therefore, this specific combination of original state, the rigidity state of sampling and bias term is designated as low good combination
Probability.Therefore, it may not be good distribution that estimated rigidity state, which is distributed with,.
In some embodiments, the generation of first state track is carried out by generating N kind state in each time step, and
And by weightIt is how good in specific time step to react the particular state to become particle with generated every kind of state relation
Predict measured value well.In some embodiments, whenever particle be designated lower than some threshold value low weight when, by particle from
It removes in estimation and is replaced with the particle with more Gao Quanchong (that is, the probability for becoming good particle is higher).Rigidity
The update of state distribution can be accomplished in several ways.One embodiment takes the state from the estimation at moment to the end of the second moment
Track and the difference from the first time step predicted motion of the state trajectory of time step to the end, use the difference to update the ginseng of distribution
Number.For example, the distribution of rigidity state is modeled as positive and negative Wishart distribution in an embodiment in.
Fig. 6 A, which is shown, generates first state track three times when for the rigidity value of each grey iterative generation five samplings
The rough schematic view of the result of iteration.Use motion model and five to the input of system and for parameterizing dynamic model
The rigidity value of sampling, in time 611a forward prediction original state 710a, with generate next five kinds of state 621a, 622a,
623a, 624a and 625a.Through determination, probability is the biasing of the model and measured value 626a with measured value 626a and noise source
The variation of 627a and change.In each time step, that is, in each iteration, generate polymerization state using the set of probability
620a。
The possibility of five kinds of states specifies probability when Fig. 6 B shows first time iteration in Fig. 6 A.Exemplar state 621b, 622b,
The selection of the size of the point of 623b, 624b and 625b reflects those probability 621b, 622b, 623b, 624b and 625b.
Can in a number of ways in any above embodiment to realize the disclosure.For example, hardware, software can be used
Or combinations thereof realize embodiment.When implemented in software, software code can be in any suitable processor or processor collection
Execution is closed, is either arranged or is dispersed throughout multiple computers in single computer.These processors can be implemented as collecting
At circuit, wherein one or more processors are in integrated circuit package.But the circuit of any suitable format can be used
To realize processor.
In addition, can be encoded as can be in using various operating systems or platform for various methods outlined herein or processing
Either one or two of one or more processors on the software that executes.In addition, using a variety of suitable programming languages and/or programming
Or any in wscript.exe writes this software, and this software can also be compiled as executing on a framework or virtual machine
Executable machine language code or intermediate code.In general, in various embodiments, can be combined by expectation or distribution program module
Function.
In addition, embodiments of the present invention can be implemented as being provided its exemplary method.Part as this method
The movement of execution can sort in any suitable way.Therefore, embodiment can be constructed, in these embodiments, with institute
Different order is illustrated to execute movement, it may include some movements are performed simultaneously, even if being shown as in illustrated embodiment
The movement of sequence.
Claims (15)
1. a kind of system for calibrating the vehicle tyre by estimating the rigidity state of vehicle tyre, the rigidity state
At least one parameter including defining the interaction of at least one tire and road of vehicle, the system include:
It is associated with the state of the vehicle to be used to store inputting to the control of the vehicle for the vehicle for memory
The measured value of the movement by the vehicle of motion model and vehicle measurement mould associated with the state of the vehicle
Type, wherein the motion model of the vehicle includes the certainty component of the movement and the probability component of the movement
Combination, wherein the certainty component of the movement is independently of the rigidity state and defines the vehicle and becomes at any time
The movement changed and changed, wherein the probability component of the movement includes with uncertainty and defining to the vehicle
The movement interference rigidity state；
Receiver is used to receive the exercise data for indicating that the vehicle moves on road according to track, wherein the movement
Data include making the vehicle according to the sequence of the mobile control input to the vehicle in the track and along the track
The sequence of the measured value of the movement of the mobile vehicle, and wherein, it is defeated that the sequence of the measured value corresponds to the control
The sequence entered；
Processor is used to iteratively update the probability distribution of the rigidity state until meeting termination condition, wherein iteration makes
One or more samples of the probability distribution of the sequence and the rigidity state that are inputted with the control are according to the motion model
It determines the first state track of the vehicle, using the sequence of the measured value determines the vehicle according to the measurement model
The second state trajectory, and update the probability distribution of the rigidity state to reduce the first state rail of the vehicle
Error between mark and second state trajectory of the vehicle；And
Output device, for rendering the probability distribution of the rigidity state with when meeting the termination condition rigidity shape
At least one of sample of probability of state distribution or combinations thereof.
2. system according to claim 1, wherein the measurement model of the vehicle includes independently of the rigidity shape
The combination of the probability component of the certainty component of the measurement model of state and the measurement model including the rigidity state,
And described wherein, is determined according to the measurement model using the sample of the sequence of the measured value and the rigidity state
Two-state track.
3. system according to claim 1, wherein each respective value with the measured value is inputted for the control, from
Sample is extracted in the probability distribution of the rigidity state.
4. system according to claim 1, wherein the control input includes the steering angle for specifying the wheel of the vehicle
With the order of the value of one of the speed of rotation of wheel or combinations thereof, and wherein, the measured value includes the vehicle
The value of one of acceleration of the speed of rotation and the vehicle or combinations thereof, and wherein, the state trajectory includes state
Sequence, every kind of state all includes the rate and course angular speed of the vehicle, so that the motion model is logical in continuous time step
The value of input is controlled described in the first value for crossing the state of the dynamic vehicle of the vehicle, and the measurement model is same
Time step is associated with the second value of the state of the vehicle by the value of the measured value.
5. system according to claim 1, wherein the processor updates the institute of the rigidity state by following steps
State probability distribution:
The error is determined as the first state track of the vehicle and second state trajectory of the vehicle
Weighted difference between corresponding states；
The value of the sample of the probability distribution of the rigidity state is adjusted, so that the error reduces；And
The probability distribution of the rigidity state is updated, so as to mention from the probability distribution of the rigidity state of update
The probability of the value of the sample of adjusting is taken to increase.
6. system according to claim 1, wherein the processor is configured for:
Determine that one group of particle, each particle represent true with the different samples extracted from the probability distribution of the rigidity state
The fixed first state track；
It will be compared from described group of each particle with second track, the particle and described second represented with determination
The weight of each particle of error between state trajectory；And
The combination of the particle that the first state track is determined as the weight according to each particle and is weighted.
7. system according to claim 1, wherein the processor is configured for:
Determine one group of initial first state corresponding from the different samples extracted from the probability distribution of the rigidity state
Track；
It will be compared from described group of each initial first trajectory with second track, represent described initial the to determine
The weight of the initial first state track of each of error between one state trajectory and second state trajectory；And
The first state track is determined as to the original state of the Weight according to each initial first state track
The combination of track.
8. a kind of method for calibrating the vehicle tyre by estimating the rigidity state of vehicle tyre, the rigidity state
At least one parameter including defining the interaction of at least one tire and road of the vehicle, wherein the method makes
With the processor coupled with the instruction of the realization the method stored, wherein described instruction by the processor when being executed
At least some steps of the method are executed, the step includes:
Receive the vehicle from memory will input associated with the state of vehicle movement to the control of the vehicle
The measured value of the movement by the vehicle of model and vehicle measurement model associated with the state of the vehicle,
In, the motion model of the vehicle includes the combination of the certainty component of the movement and the probability component of the movement,
Wherein, the certainty component of the movement independently of the rigidity state and defines the vehicle and changes over time and become
The movement of change, wherein the probability component of the movement includes with uncertain and definition to described in the vehicle
The rigidity state of the interference of movement；
Receive the exercise data for indicating that the vehicle moves on road according to track, wherein the exercise data includes making institute
Vehicle is stated according to the sequence of the mobile control input to the vehicle in the track and the vehicle moved along the track
Movement measured value sequence, and wherein, the sequence of the measured value corresponds to the sequence of the control input；
The probability distribution of the rigidity state is iteratively updated until meeting termination condition, wherein iteration is defeated using the control
One or more samples of the probability distribution of the sequence and the rigidity state that enter determine the vehicle according to the motion model
First state track, the second state of the vehicle is determined according to the measurement model using the sequence of the measured value
Track, and update the probability distribution of the rigidity state come reduce the vehicle the first state track and the vehicle
Second state trajectory between error；And
The probability distribution and the probability distribution of the rigidity state when meeting the termination condition that the rigidity state is presented
At least one of sample or combinations thereof.
9. according to the method described in claim 8, wherein, the measurement model of the vehicle includes independently of the rigidity shape
The combination of the probability component of the certainty component of the measurement model of state and the measurement model including the rigidity state,
And described wherein, is determined according to the measurement model using the sample of the sequence of the measured value and the rigidity state
Two-state track.
10. according to the method described in claim 8, wherein, for each respective value of the control input and the measured value,
Sample is extracted from the probability distribution of the rigidity state.
11. according to the method described in claim 8, wherein, the control input includes specifying the steering of the wheel of the vehicle
The order of the value of one of speed of rotation of angle and the wheel or combinations thereof, and wherein, the measured value includes described
The value of one of acceleration of the speed of rotation of vehicle and the vehicle or combinations thereof, and wherein, the state trajectory packet
The sequence of state is included, every kind of state all includes the rate and course angular speed of the vehicle, so that the motion model is continuous
The value of input, and the measurement model are controlled described in the first value of the time step by the state of the dynamic vehicle of the vehicle
It is in same time step that the value of the measured value is associated with the second value of the state of the vehicle.
12. according to the method described in claim 8, wherein, the probability distribution for updating the rigidity state includes following step
It is rapid:
The error is determined as the first state track of the vehicle and second state trajectory of the vehicle
Weighted difference between corresponding states；
The value of the sample of the probability distribution of the rigidity state is adjusted, so that the error reduces；And
The probability distribution of the rigidity state is updated, so as to mention from the probability distribution of the rigidity state of update
The probability of the value of the sample of adjusting is taken to increase.
13. according to the method described in claim 8, the method also includes following steps:
Determine that one group of particle, each particle represent true with the different samples extracted from the probability distribution of the rigidity state
The fixed first state track；
It will be compared from described group of each particle with second track, the particle and described second represented with determination
The weight of each particle of error between state trajectory；And
The combination of the particle that the first state track is determined as the weight according to each particle and is weighted.
14. according to the method described in claim 8, the method also includes following steps:
Determine one group of initial first state corresponding from the different samples extracted from the probability distribution of the rigidity state
Track；
It will be compared from described group of each initial first trajectory with second track, represent described initial the to determine
The weight of the initial first state track of each of error between one state trajectory and second state trajectory；And
The first state track is determined as to the original state of the Weight according to each initial first state track
The combination of track.
15. a kind of non-transient computer readable storage medium of implementation procedure on it, described program can be executed by processor with
Just a kind of method is executed, method includes the following steps:
From memory reception vehicle motion model associated with the state of the vehicle will be inputted to the control of the vehicle
And the measured value measurement model associated with the state of the vehicle of the movement by the vehicle of the vehicle, wherein
The motion model of the vehicle includes the combination of the certainty component of the movement and the probability component of the movement,
In, the certainty component of the movement is independently of rigidity state and defines the fortune that the vehicle changes over time and change
It is dynamic, wherein the probability component of the movement includes with the movement uncertain and that definition is to the vehicle
The rigidity state of interference；
Receive the exercise data for indicating that the vehicle moves on road according to track, wherein the exercise data includes making institute
Vehicle is stated according to the sequence of the mobile control input to the vehicle in the track and the vehicle moved along the track
Movement measured value sequence, and wherein, the sequence of the measured value corresponds to the sequence of the control input；
The probability distribution of the rigidity state is iteratively updated until meeting termination condition, wherein iteration is defeated using the control
One or more samples of the probability distribution of the sequence and the rigidity state that enter determine the vehicle according to the motion model
First state track, the second state of the vehicle is determined according to the measurement model using the sequence of the measured value
Track, and update the probability distribution of the rigidity state come reduce the vehicle the first state track and the vehicle
Second state trajectory between error；And
The probability distribution and the probability distribution of the rigidity state when meeting the termination condition that the rigidity state is presented
At least one of sample or combinations thereof.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US15/467,141 | 2017-03-23 | ||
| US15/467,141 US10093321B1 (en) | 2017-03-23 | 2017-03-23 | System and method for calibrating parameters of tires |
| PCT/JP2017/035082 WO2018173340A1 (en) | 2017-03-23 | 2017-09-21 | System and method for calibrating tire of vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110475697A true CN110475697A (en) | 2019-11-19 |
| CN110475697B CN110475697B (en) | 2021-06-11 |
Family
ID=60164759
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201780088545.4A Active CN110475697B (en) | 2017-03-23 | 2017-09-21 | System and method for calibrating vehicle tires |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US10093321B1 (en) |
| EP (1) | EP3600985B1 (en) |
| JP (1) | JP6815519B2 (en) |
| CN (1) | CN110475697B (en) |
| WO (1) | WO2018173340A1 (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6748619B2 (en) * | 2017-09-20 | 2020-09-02 | 日立オートモティブシステムズ株式会社 | Vehicle control device, vehicle control method, and vehicle control system |
| EP3501944B1 (en) * | 2017-12-20 | 2020-08-05 | Aptiv Technologies Limited | Method and device for estimating a steering torque |
| US10408638B2 (en) * | 2018-01-04 | 2019-09-10 | Mitsubishi Electric Research Laboratories, Inc. | System and method for controlling a vehicle under sensor uncertainty |
| KR102644549B1 (en) * | 2018-10-29 | 2024-03-07 | 현대자동차주식회사 | Integrated control system for vehicle |
| AU2019396213A1 (en) * | 2018-12-11 | 2021-05-27 | SafeAI, Inc. | Techniques for kinematic and dynamic behavior estimation in autonomous vehicles |
| US11597364B2 (en) * | 2019-03-11 | 2023-03-07 | Mitsubishi Electric Research Laboratories, Inc. | System and method for determining friction curve of tire |
| US11254315B2 (en) * | 2019-03-12 | 2022-02-22 | Mitsubishi Electric Research Laboratories, Inc. | Friction adaptive vehicle control |
| EP3983861B1 (en) * | 2019-06-14 | 2023-10-04 | Volvo Truck Corporation | A method for quantifying correctness of a vehicle model |
| US11359919B2 (en) * | 2019-10-04 | 2022-06-14 | Aptiv Technologies Limited | Calculating vehicle states of a vehicle system for lane centering |
| US11084471B2 (en) * | 2019-11-22 | 2021-08-10 | Automotive Research & Testing Center | Braking control method according to friction of road surface |
| US11554779B2 (en) * | 2019-11-25 | 2023-01-17 | Hyundai Mobis Co., Ltd. | Apparatus and method for controlling backward driving of vehicle |
| US11203354B2 (en) * | 2020-02-26 | 2021-12-21 | Mitsubishi Electric Research Laboratories, Inc. | System and method for determining friction curve of tire |
| CN111368424B (en) * | 2020-03-03 | 2023-09-01 | 阿波罗智能技术(北京)有限公司 | Vehicle simulation method, device, equipment and medium |
| CN112434404A (en) * | 2020-11-07 | 2021-03-02 | 西南交通大学 | Turnout rigid-flexible coupling vibration analysis method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040225423A1 (en) * | 2003-05-07 | 2004-11-11 | Carlson Christopher R. | Determination of operational parameters of tires in vehicles from longitudinal stiffness and effective tire radius |
| GB2461551A (en) * | 2008-07-03 | 2010-01-06 | Gm Global Tech Operations Inc | Estimating the sideslip velocity of a vehicle. |
| CN103921778A (en) * | 2013-01-10 | 2014-07-16 | 英飞凌科技股份有限公司 | Wheel localizer, wheel localization device, system, and method for locating a position of a wheel |
| EP3028909A1 (en) * | 2014-12-03 | 2016-06-08 | The Goodyear Tire & Rubber Company | Intelligent tire-based road friction estimation system and method |
| EP3059130A1 (en) * | 2015-02-20 | 2016-08-24 | Volvo Car Corporation | Method, arrangement and system for estimating vehicle cornering stiffness |
| EP3115765A1 (en) * | 2015-07-08 | 2017-01-11 | The Goodyear Tire & Rubber Company | Tire sensor-based vehicle state estimation system and method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2835919A1 (en) * | 2002-02-08 | 2003-08-15 | Michelin Soc Tech | MEASUREMENT OF MAXIMUM ADHESION COEFFICIENT FROM KNOWLEDGE OF GENERATED EFFORTS AND SELF-ALIGNMENT TORQUE IN THE TIRE CONTACT AIR |
| JP2007153034A (en) * | 2005-12-01 | 2007-06-21 | Toyota Motor Corp | Tire abrasion state judging device |
| DE102006054805B4 (en) * | 2006-11-21 | 2009-11-26 | Continental Automotive Gmbh | Method and apparatus for estimating a friction index |
| ATE528185T1 (en) * | 2007-03-01 | 2011-10-15 | Haldex Brake Prod Ab | SYSTEMS AND METHODS FOR DETERMINING A PARAMETER RELATED TO THE CONTACT BETWEEN TIRE AND ROAD AND/OR THE RELATIONSHIP BETWEEN A WHEEL AND VEHICLE MOTION |
| JP5707790B2 (en) * | 2010-09-06 | 2015-04-30 | 日産自動車株式会社 | Tire contact state estimation device |
-
2017
- 2017-03-23 US US15/467,141 patent/US10093321B1/en active Active
- 2017-09-21 JP JP2019534772A patent/JP6815519B2/en active Active
- 2017-09-21 EP EP17788316.2A patent/EP3600985B1/en active Active
- 2017-09-21 CN CN201780088545.4A patent/CN110475697B/en active Active
- 2017-09-21 WO PCT/JP2017/035082 patent/WO2018173340A1/en unknown
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040225423A1 (en) * | 2003-05-07 | 2004-11-11 | Carlson Christopher R. | Determination of operational parameters of tires in vehicles from longitudinal stiffness and effective tire radius |
| GB2461551A (en) * | 2008-07-03 | 2010-01-06 | Gm Global Tech Operations Inc | Estimating the sideslip velocity of a vehicle. |
| CN103921778A (en) * | 2013-01-10 | 2014-07-16 | 英飞凌科技股份有限公司 | Wheel localizer, wheel localization device, system, and method for locating a position of a wheel |
| EP3028909A1 (en) * | 2014-12-03 | 2016-06-08 | The Goodyear Tire & Rubber Company | Intelligent tire-based road friction estimation system and method |
| EP3059130A1 (en) * | 2015-02-20 | 2016-08-24 | Volvo Car Corporation | Method, arrangement and system for estimating vehicle cornering stiffness |
| EP3115765A1 (en) * | 2015-07-08 | 2017-01-11 | The Goodyear Tire & Rubber Company | Tire sensor-based vehicle state estimation system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| EP3600985A1 (en) | 2020-02-05 |
| US10093321B1 (en) | 2018-10-09 |
| WO2018173340A1 (en) | 2018-09-27 |
| JP2020504049A (en) | 2020-02-06 |
| US20180273046A1 (en) | 2018-09-27 |
| CN110475697B (en) | 2021-06-11 |
| JP6815519B2 (en) | 2021-01-20 |
| EP3600985B1 (en) | 2020-08-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110475697A (en) | System and method for calibrating vehicle tyre | |
| CN109476287A (en) | Method for controlling a vehicle and system | |
| CN112758097B (en) | State prediction and estimation method for unmanned vehicle | |
| CN101655504B (en) | Vehicle speed estimation method of motor vehicle self-adaption cruise system | |
| JP7475554B2 (en) | Stochastic model predictive control of uncertain systems. | |
| Lian et al. | Cornering stiffness and sideslip angle estimation based on simplified lateral dynamic models for four-in-wheel-motor-driven electric vehicles with lateral tire force information | |
| Chen et al. | Longitudinal-lateral-cooperative estimation algorithm for vehicle dynamics states based on adaptive-square-root-cubature-Kalman-filter and similarity-principle | |
| JP7285971B2 (en) | System and method for determining tire friction curve | |
| US12084073B2 (en) | Method and device for optimum parameterization of a driving dynamics control system for vehicles | |
| CN113165614B (en) | Tire stiffness estimation and road friction estimation | |
| Mazzilli et al. | On the benefit of smart tyre technology on vehicle state estimation | |
| CN108058709A (en) | For running the method and apparatus of driver assistance system, driver assistance system | |
| Hermansdorfer et al. | A concept for estimation and prediction of the tire-road friction potential for an autonomous racecar | |
| CN114435371B (en) | Road gradient estimation method and device | |
| Ye et al. | Estimation of tire-road friction coefficient with adaptive tire stiffness based on RC-SCKF | |
| CN116588119B (en) | Vehicle state estimation method based on tire model parameter self-adaption | |
| Ehlers et al. | Optimized Tuning of an EKF for State and Parameter Estimation in a Semitrailer | |
| CN111703429B (en) | Method for estimating longitudinal speed of wheel hub motor driven vehicle | |
| EP4110663B1 (en) | System and method for determining friction curve of tire | |
| JPWO2020183871A5 (en) | ||
| Schäfke et al. | Transformer Neural Networks for Maximum Friction Coefficient Estimation of Tire-Road Contact using Onboard Vehicle Sensors | |
| CN117104247A (en) | Method and system for estimating vehicle speed of bicycle and computer readable storage medium | |
| CN115071727A (en) | Method for determining speed of curve, intelligent driving simulation test system and medium | |
| CN118182485A (en) | Automobile road surface adhesion coefficient estimation method and device, storage medium and automobile | |
| CN118560525A (en) | Vehicle transverse control method and device, vehicle and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |